Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3287 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3588. Min Act: -0.0001

Data Index: 905242 (C4 (Web Text))

Max Activating Token Index: 576

Text #1

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3662. Min Act: -0.0001

Data Index: 309037 (C4 (Web Text))

Max Activating Token Index: 48

Text #2

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3628. Min Act: -0.0001

Data Index: 528934 (C4 (Web Text))

Max Activating Token Index: 338

Text #3

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3182. Min Act: -0.0001

Data Index: 1527324 (Python Code)

Max Activating Token Index: 409

Text #4

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3226. Min Act: -0.0001

Data Index: 292796 (C4 (Web Text))

Max Activating Token Index: 832

Text #5

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3365. Min Act: -0.0001

Data Index: 423181 (C4 (Web Text))

Max Activating Token Index: 745

Text #6

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3322. Min Act: -0.0001

Data Index: 1265043 (C4 (Web Text))

Max Activating Token Index: 453

Text #7

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3426. Min Act: -0.0001

Data Index: 160643 (C4 (Web Text))

Max Activating Token Index: 275

Text #8

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3271. Min Act: -0.0001

Data Index: 26659 (C4 (Web Text))

Max Activating Token Index: 496

Text #9

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3543. Min Act: -0.0001

Data Index: 1048624 (C4 (Web Text))

Max Activating Token Index: 108

Text #10

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3231. Min Act: -0.0001

Data Index: 152560 (C4 (Web Text))

Max Activating Token Index: 753

Text #11

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3462. Min Act: -0.0001

Data Index: 185344 (C4 (Web Text))

Max Activating Token Index: 643

Text #12

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3443. Min Act: -0.0001

Data Index: 858037 (C4 (Web Text))

Max Activating Token Index: 370

Text #13

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3348. Min Act: -0.0001

Data Index: 1234522 (C4 (Web Text))

Max Activating Token Index: 80

Text #14

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3179. Min Act: -0.0001

Data Index: 101587 (C4 (Web Text))

Max Activating Token Index: 873

Text #15

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3117. Min Act: -0.0001

Data Index: 191379 (C4 (Web Text))

Max Activating Token Index: 310

Text #16

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3123. Min Act: -0.0001

Data Index: 16441 (C4 (Web Text))

Max Activating Token Index: 382

Text #17

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.2941. Min Act: -0.0001

Data Index: 211130 (C4 (Web Text))

Max Activating Token Index: 303

Text #18

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3013. Min Act: -0.0001

Data Index: 1085296 (C4 (Web Text))

Max Activating Token Index: 745

Text #19

Max Range: 0.3662. Min Range: -0.3662

Max Act: 0.3254. Min Act: -0.0001

Data Index: 1195022 (C4 (Web Text))

Max Activating Token Index: 860